Adoptive transfer of autologous T cells engineered to express chimeric antigen receptors (CARs) has emerged as a promising immunotherapy for patients with hematologic malignancies, such as CD19 CARs in leukemias and lymphomas. However, challenges remain in terms of manufacturing consistency and the functional profile of the CAR-T cell product, since infused cells, as ?living drugs?, can be activated and release a variety of cytokines upon specific antigen recognition. These secreted cytokines may result in not only therapeutic efficacy, but also life-threatening immunotoxicity, such as cytokine release syndrome and neurologic toxicity. It is essential for cancer patients to have a full spectrum potency and toxicity profiling of CAR-T cell products before infusion. We have developed a single-cell barcode chip (SCBC) technology, which enables (a) assaying 30+ secreted proteins per individual live cell, (b) a broad range of immune cell functions covering efficacy and safety, (c) is designed to fit various types of immune cells, (d) is the first quantitative metric of these proteins per cell, and (e) requires samples sizes totaling only 5000 cells, all critical leaps over flow cytometry based platforms. With the collaboration of two leading CAR-T pharmaceutical companies, IsoPlexis? single-cell deep profiling has revealed, for the first time, a pre-infusion correlate to post-infusion patient response. The SCBC analysis of 20 patients with lymphoma demonstrates a significant correlation of the polyfunctional strength of the CAR-T cells with objective response (complete or partial) to the CAR-T cell therapy (p = 0.012), indicating a powerful metric of quality assessment, where the other gold standard technologies did not detect any significant correlations. Based on the clinical data and the automation progress in our SBIR Phase I grant, we are proposing the following specific aims in this two-year Phase II application: AIM 1: Develop automated ?flow cell? consumable compartment, which captures 32-plex single-cell cytokine response, to prepare for full automation required by large CAR-T trial (months 1-10); AIM 2: Produce a fully automated cartridge analysis and workflow system, to allow 10 samples assayed in parallel with minimal user interaction, allowing easy introduction into clinical core labs (months 1-15); AIM 3: Produce a comprehensive informatics suite, and fully test our automated system and informatics in two 30 patient cohort trials, providing robust and predictive biomarkers for pre-infusion CAR-T quality. Establish MSKCC and UCLA beta sites for post-phase II transition (months 7-24). With this Phase II submission, we will deliver the first effective CAR-T pre-infusion quality check assay to predict objective response in patients, in an easy-to-use, automated system that can be used throughout all cellular immunotherapy trials.